by Loopera-ai
面向基本面因子研究的智能体-A hypothesis-driven AI agent for fundamental factor research, with evidence-gated validation and research memory.
# Add to your Claude Code skills
git clone https://github.com/Loopera-ai/looperaSee how loopera compares with popular alternatives.
loopera is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by Loopera-ai. 面向基本面因子研究的智能体-A hypothesis-driven AI agent for fundamental factor research, with evidence-gated validation and research memory. It has 71 GitHub stars.
loopera's catalog security scan is still queued. You can run an instant dependency and prompt-injection check now with the "Scan for vulnerabilities" button above.
Clone the repository with "git clone https://github.com/Loopera-ai/loopera" and add it to your Claude Code skills directory (see the Installation section above).
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh loopera against similar tools.
No comments yet. Be the first to share your thoughts!
Unlocks once the catalog security scan passes (runs nightly).
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
The deep catalog scan for this skill is still queued. Run an instant dependency check now instead.
仓库定位: 这是 Loopera 的公开产品与技术资料仓库,不是源代码发行包。当前不提供可运行代码、安装包、CLI 或公开 API;README 中的流程和案例用于说明产品能力与证据边界。
Loopera 围绕上市公司的财务报表与经营信息,为基本面量化团队持续完成研究方向探索、经济假设形成、候选因子构建、可信验证与知识沉淀。
| Research | Validate | Remember | Collaborate |
|---|---|---|---|
| 从跨报表经营关系中发现值得解释的现象 | 分离研究逻辑与历史表现,逐层筛选候选 | 保存成果、失败原因与研究谱系 | 支持研究员在关键节点审阅、干预和决策 |
| 形成可解释、可检验的基本面假设 | 检查时序、口径、稳定性、重复度与增量价值 | 避免重复试错,持续发现研究空白 | 适配小规模任务与多角色协作研究 |
它的目标不是批量生成公式,也不是用一次漂亮的回测替代研究判断。Loopera 更关心一项研究是否有清晰的基本面逻辑、是否经得住多角度验证,以及它是否为已有研究带来新的信息。
Loopera 将基本面研究从一次性的人工尝试,组织成持续运行、能够积累和进化的研究流程。
价格反映结果,基本面解释企业如何创造、消耗和重新配置价值。
Loopera 主要围绕企业的盈利质量、现金流、资产负债结构、营运效率、资本投入和经营变化展开研究。相比只观察市场价格,它更关注企业经营中正在发生、但尚未被充分理解或定价的变化。
基本面数据并不简单。不同报表之间需要建立联系,同一个财务数字在不同行业和经营阶段可能有不同含义,公开信息的时间也会影响研究是否真实可用。Loopera 将这些问题纳入统一的研究流程,让 Agent 的工作更接近基本面研究员,而不是通用的数据拟合工具。
报告中的明确公式可以优先进入验证,其余内容则作为后续探索的背景知识。外部材料始终被视为待验证的研究证据,不能修改 Loopera 的安全约束、数据范围和验证标准。
flowchart LR
A["研究输入<br/>Topic · Prompt · PDF"] --> B["基本面理解<br/>字段 · 报表 · 行业语境"]
B --> C["智能研究<br/>现象 · 机制 · 候选"]
C --> D["可信验证<br/>时点 · 逻辑 · 稳健性 · 增量"]
D --> E["研究记忆<br/>成果 · 失败 · 谱系"]
E -. "反哺下一轮" .-> C
classDef input fill:#E8F7FF,stroke:#168CFF,color:#07142E,stroke-width:1.5px;
classDef research fill:#F1ECFF,stroke:#7C3AED,color:#07142E,stroke-width:1.5px;
classDef evidence fill:#E8FFF7,stroke:#0F9F8F,color:#07142E,stroke-width:1.5px;
classDef memory fill:#FFF2FB,stroke:#C13C8A,color:#07142E,stroke-width:1.5px;
class A,B input;
class C research;
class D evidence;
class E memory;
产品闭环的终点不是一条因子,而是可复核的研究资产。 每次运行都保留问题、假设、证据、结论与后续方向。
公开版本仅介绍产品能力和研究原则。具体的假设组织方式、验证策略、评价规则、模型协作机制和因子实现属于 Loopera 的核心技术,不在本文档中披露。
flowchart LR
A["AI Agent<br/>提出 · 解释 · 修订"] --> B["Research Contract<br/>假设 · 字段 · 方向 · 约束"]
B --> C["Deterministic Harness<br/>计算 · 检查 · 评价"]
C --> D["Research Memory<br/>结果 · 失败 · 谱系"]
D -. "提供研究上下文" .-> A
H["Human Researcher<br/>审阅 · 干预 · 决策"] <--> B
classDef agent fill:#F1ECFF,stroke:#7C3AED,color:#07142E,stroke-width:1.5px;
classDef contract fill:#E8F7FF,stroke:#168CFF,color:#07142E,stroke-width:1.5px;
classDef evidence fill:#E8FFF7,stroke:#0F9F8F,color:#07142E,stroke-width:1.5px;
classDef memory fill:#FFF2FB,stroke:#C13C8A,color:#07142E,stroke-width:1.5px;
classDef human fill:#FFF8E6,stroke:#D98E04,color:#07142E,stroke-width:1.5px;
class A agent;
class B contract;
class C evidence;
class D memory;
class H human;
flowchart TB
subgraph L1["01 · 研究交互与编排层"]
direction LR
A["研究输入与材料理解"] --> B["Agent 编排"] --> C["轨迹与报告"]
end
subgraph L2["02 · 假设与因子研究层"]
direction LR
D["假设研究"] --> E["因子构建"] --> F["可信验证"]
end
subgraph L3["03 · 数据、评价与记忆层"]
direction LR
G["数据与字段系统"] --> H["回测与增量评估"] --> I["研究记忆与规划"]
end
A --> D
B --> D
E --> H
F --> H
H --> C
G --> D
G --> E
I -. "反馈研究上下文" .-> D
classDef experience fill:#E8F7FF,stroke:#168CFF,color:#07142E,stroke-width:1.5px;
classDef engine fill:#F1ECFF,stroke:#7C3AED,color:#07142E,stroke-width:1.5px;
classDef foundation fill:#E8FFF7,stroke:#0F9F8F,color:#07142E,stroke-width:1.5px;
class A,B,C experience;
class D,E,F engine;
class G,H,I foundation;